Increasingly, the Democratic primary appears to be a two-person race — just in time for Super Tuesday.
There hasn’t been a ton of polling in some of Super Tuesday’s 14 states and one territory — while two polls were released for Alabama on Tuesday, for example, its last poll before that was conducted in July 2019. But based on the polling that does exist, as well as national polling and what we know about state and demographic trends, it would appear that Biden is poised to win in six or seven states, and Sanders is likely to take the top spot in seven to nine contests.
For instance, a national Morning Consult online poll taken March 1 — after the South Carolina primary — found Sanders to have a 3 percentage point lead on Biden, with the senator having 29 percent support and the former vice president 26 percent. Former New York City Mayor Mike Bloomberg was third and Sen. Elizabeth Warren fourth, with 17 and 11 percent support, respectively. The poll, which surveyed 2,656 likely Democratic primary voters, has a margin of error of 2 percentage points, meaning Biden could actually be first and Sanders second.
These results are somewhat disappointing for Bloomberg and Warren, both of whom have made big Super Tuesday bets. Bloomberg did not compete in the race’s first four contests, believing instead that his massive advertising campaign — on which he’s spent more than half a billion dollars — would help him to net a number of delegates. And Warren’s campaign, which has struggled in the race’s early states, said in February it was planning for “a drawn-out contest to accumulate delegates everywhere.”
How well these strategies work remains to be seen, but polling suggests Bloomberg is unlikely to win any states outright; Warren is running a close race with Sanders in her home state of Massachusetts and may win there — otherwise, she is in the running for delegates in a number of other states, including California, Colorado, and Maine.
But the big question is how many delegates Sanders and Biden will each receive, and whether either candidate will get enough to place him in the running to win a majority of delegates before July’s Democratic National Convention, thus avoiding the drama a contested convention would bring.
To answer that question, we have to look at what the polls tell us about key races, what we know about certain states’ voting patterns, and how the late candidate exits could affect the race.
Could a Biden surge put him back on par with Sanders nationally?
Last-minute polling for most states was released Tuesday, with most states seeing one or two new polls. Many of those states, however, like Tennessee and Alabama, hadn’t been polled in more than six months, making it difficult to tell whether these new polls fit into a clear pattern or are outliers. When in doubt — as we are in quite a few Super Tuesday states, despite the 11th-hour polls — it’s a good idea to look to the national polling.
And in that, it’s clear that Sanders and Biden are at the top of the pack.
The Morning Consult result is the only major national work available that was produced in the wake of the South Carolina primary, which gave Biden a huge win — he finished there with 48 percent of the vote — and one that appears to have changed the dynamics of the race in a significant way. Three candidates (Sen. Amy Klobuchar, former South Bend, Indiana, Mayor Pete Buttigieg, and entrepreneur Tom Steyer) dropped out after a disappointing showing there. And both Klobuchar and Buttigieg have endorsed Biden.
Because of this, none of the national polls we have are as accurate as they could be — few are available that attempt to measure the effect South Carolina’s results have had on the race, and none omitted Steyer, Buttigieg, or Klobuchar as candidate choices.
But even with this said, it is clear the latest Morning Consult poll is of a kind with other recent polling: An Investment Business Daily/TIPP poll of 325 registered voters fielded from February 20 to 29 found Sanders leading Biden by 3 percentage points (with a 3.3 percentage point margin of error), and a Yahoo News/YouGov poll of 1,662 registered voters taken from February 26 to 27 (with a 3 percentage point margin of error) found Sanders to have a 7-point lead.
These national polls are instructive in states where there has been a dearth of recent polling, like Alabama, a state with 52 delegates on offer. In the month before the New Hampshire primary, more than 20 polls were released. But in the seven months before Super Tuesday, zero Alabama polls were produced. The two that we now have were both released Tuesday. One of those, from a new polling platform called Swayable, had results its pollsters caveated by saying they were released amid “ongoing work refining population and voter models of the platform.” The other comes from the progressive think tank Data for Progress, which has had a good track record in 2020 so far.
This reality makes the national polls helpful. And those suggest that Biden’s level of support has likely risen following his South Carolina performance — for instance, Morning Consult’s national pre- and post-South Carolina polling found a 7 percentage point boost in support for Biden (and a 3-point loss for Sanders). And the latest state polling suggests the same. Taking Alabama again as an example, the Data for Progress work (a text message and online survey of 237 likely voters) was taken in the two days following South Carolina, and found Biden to have a 25-point lead on Sanders, his closest rival, with a 6.4 percentage point margin of error.
Alabama was always a state that favored Biden given its demographics. But eleventh-hour polls from Swayable and Data for Progress also show a major Biden bump in states that were not seen as strongholds for him, like Texas.
Again, these most recent polls are far from definitive. But they, along with the recent national polling we have, strongly suggest Sanders’s status as the primary’s frontrunner is being seriously challenged by a surging Biden, and that it is the former vice president who may finish at the top of the pack Tuesday. So much so, in fact, that forecasters at FiveThirtyEight nearly doubled Biden’s chances of winning the nomination on the first ballot Tuesday morning.
California and Texas are the day’s most delegate-rich states
California and Texas have about 48 percent of Super Tuesday’s delegate total, and 16 percent of all the pledged delegates in the primary.
This makes winning both states key to any true Super Tuesday victory, and in both, Sanders is a clear frontrunner.
In California, he has consistently topped polls since last December, with Biden or Warren usually coming in second. Recent polls have recorded a growing Sanders lead, with a February 26-29 USA Today poll finding him 19 percentage points ahead of Biden, and a February 29-March 1 Emerson College poll finding Sanders to have a 17-point lead on the former vice president. Both leads are, of course, strong enough to fall outside the polls’ margins of error (5.7 and 4.1 percentage points, respectively).
The polls released Tuesday from Swayable and Data for Progress suggest Biden may be cutting into Sanders’s lead but that it remains large enough for the senator’s victory to be all but assured.
The Data for Progress poll — one of three California polls taken post-South Carolina — has the strongest result for Biden in weeks; pollsters reported his support at 25 percent to Sanders’s 32 percent. Bloomberg was third with 17 percent support, and Warren was fourth, with 16 percent.
All this signals that Sanders is in good shape to win a large share of California’s 415 delegates. But Biden could win nearly as large a share. And both Bloomberg and Warren could leave California with impressive delegate hauls too.
But things could be much wackier.
There are three connected reasons for this: The Democratic Party awards delegates proportionately to candidates who clear a certain threshold of votes, California has a robust early voting program, and the candidates who dropped out could have a significant impact on the results.
As Vox’s Andrew Prokop has explained, delegates are awarded only to candidates who receive more than 15 percent of the vote, either statewide or in a state’s congressional districts. This means that should Bloomberg’s California results reflect, say, Emerson College’s polling rather than Data for Progress’s (at both the statewide level and in each district), he wouldn’t get any delegates, as Emerson’s pollsters found him receiving 11 percent support.
Instead, again assuming the results reflect Emerson’s work, Sanders, Biden, and Warren (who reached 16 percent support in the Emerson poll) would split all the 415 delegates between themselves proportionately, based on the percentage of the remaining vote they received.
But even if the results are more like Emerson’s work than more recent polls, Bloomberg could still win some of the 272 district-level delegates on offer if that 11 percent support is concentrated in just a few congressional districts.
Moreover, all the candidates’ final delegate shares will likely be different from those suggested by polling, because Buttigieg, Steyer, and Klobuchar are no longer on the ballot — even the most recent polls included these candidates as options.
How different is not clear, because California allows for early voting, and more than 2.7 million people had submitted their ballots ahead of Super Tuesday. Those ballots that were for the three candidates who dropped out are locked in. But those who waited until Tuesday will be able to vote for whoever their second choice is.
Buttigieg, Steyer, and Klobuchar didn’t have massive support in the state, but the Emerson poll found them to have about 14 percent of the vote between them, enough to influence viability, particularly if the majority were to migrate to one candidate.
A similar situation exists in Texas. There, polls suggest a very close race between Sanders and Biden. Emerson’s February 29-March 1 Texas poll found Sanders leading Biden by 5 percentage points (31 percent support for Sanders compared to 26 percent for Biden). The poll has a 4.6 percentage point margin of error, meaning the race could be even closer than its results suggest — a reading supported by the Swayable and Data for Progress March 1-2 polls, which, again, have the candidates tied.
Potentially affecting all this is the fact Biden also received an endorsement from popular Texas politico and former 2020 candidate former Rep. Beto O’Rourke on Monday. How much endorsements matter has long been a question in politics, but at least one has made a difference for Biden recently: that of South Carolina Rep. Jim Clyburn, who endorsed Biden just days ahead of voting in South Carolina. An Edison Research exit poll of that state found 61 percent of voters said Clyburn’s endorsement influenced their vote, and Biden earned a resounding victory there. A similar paradigm in Texas with O’Rourke could tip the scales in Biden’s favor.
Texas also allows early voting, and at least 690,000 people voted prior to Tuesday. This means that, as in California, Biden might not receive a massive boost in support due to the endorsements. He should, at the very least, leave Texas with a delegate total similar to Sanders, however.
Biden is well positioned to sweep the South
There are a number of states that, despite there being inconsistent polling ahead of Tuesday, Biden is expected to carry. All of these states — Alabama, Arkansas, Tennessee, North Carolina, and Virginia — are in the South.
Exit polls have shown there are two groups with which Biden tends to do overwhelmingly well: black voters and older voters. These states play to those strengths.
In South Carolina, for instance, a Washington Post exit poll found Biden received 61 percent of the black vote, placing him 44 percentage points ahead of his nearest competitor for that demographic, Sanders. The Post’s entrance poll for Nevada found he won among black voters there, too, with 38 percent of the vote.
Some national polls have begun to show an erosion of Biden’s black support, like a Reuters/Ipsos study that found Sanders surging to lead among black voters, with 26 percent support to Biden’s 23 percent. It’s important to note, however, that this poll (which had a confidence interval of 2 to 5 percentage points) was taken before South Carolina, from February 19 to 25; many others that found similar results were taken around the same period.
Even if Biden’s black support is weaker than it was, he does appear to have maintained his ability to win among older voters by similar margins. In South Carolina, he carried voters over 65 by 52 percentage points; in Nevada and New Hampshire, he won that group by 12 percentage points.
The win in South Carolina was important for Biden not just because it changed the narrative around his campaign but because it could be a harbinger of contests to come. As my colleague Li Zhou has explained, “Historically, at least four Southern states — Alabama, Louisiana, Georgia, and Mississippi — have voted for the same Democratic nominee as South Carolina, giving this candidate a windfall of delegates.”
Only Alabama is voting this time on Super Tuesday, but the reasoning behind that windfall can be applied more broadly: A candidate who can win in South Carolina — a state with an electorate that this year was about 56 percent black, and in which 71 percent was over the age of 45 — can and should win other, similar states.
In Alabama, for instance, the 2016 primary electorate was 54 percent black and 61 percent over the age of 45. Arkansas’s Democratic electorate is 27 percent black and 65 percent over 45. A similar trend — on age, race, or both — holds for all of Super Tuesday’s Southern states as well as another, non-Southern state Biden is expected to win: Oklahoma.
This reality has given Biden a confidence he has begun to project on the campaign trail. Even ahead of his win in South Carolina, he told a crowd in North Carolina Saturday afternoon, “Today is a great day, because I tell you what, the full comeback starts in South Carolina. … We’re going to win South Carolina. And the next step is North Carolina.”
After picking up wins in both states, he claimed “it’s a straight path to the nomination for president of the United States of America.”
It may not be quite as simple as that, but polls released Tuesday suggest Biden has every reason to be optimistic about the South.
Take Data for Progress’s work: Biden’s leads in Alabama, Arkansas, and Virginia are enough to overcome those states’ margins of error — and his leads in the other Southern states, particularly in North Carolina, are nearly large enough to do so.
This means focusing on the South could pay major dividends for Biden. As Zeeshan Aleem has noted for Vox, Alabama, Arkansas, Tennessee, North Carolina, and Virginia have 356 pledged delegates on offer Tuesday; big wins in these states, plus in Oklahoma with its 37 delegates, and strong finishes in Texas and California begin to add up to a significant number of delegates — and perhaps even a Super Tuesday win.
Home states may not always provide a home-field advantage
While Biden holds an advantage with states featuring more senior and black voters, some other candidates hope to capitalize on the fact that their home states are voting Tuesday.
Before dropping out, Sen. Amy Klobuchar seemed about to have her first victory of the primary cycle in Minnesota’s primary, leading the two recent, pre-Super Tuesday polls in the state, which has 75 delegates, by 6 percentage points.
She may still see a strong result in that state — more than 57,196 Minnesota Democrats have already turned in their absentee ballots, but her exit looks to advantage Sanders, who was second in those polls, with 23 percent support in the Star Tribune’s work (with a 4.5 percentage point margin of error) and 21 percent in a UMass Lowell poll (with a 6.4 percentage point margin of error). In both polls, Biden was a distant fourth, polling at 8 percent in the Star Tribune survey and 9 percent in UMass Lowell’s.
Data for Progress’s poll realigned its respondents following Klobuchar’s exit, shifting her supporters (as well as those who supported Buttigieg and Steyer) to their second choices. After this process, with a margin of error of 3.8 percentage points, its pollsters found Sanders in first with 32 percent support, Biden in second with 27 percent, Warren in third with 21 percent, and Bloomberg in fourth with 16 percent.
This would seem to suggest Klobuchar’s endorsement had an effect on the race in Minnesota, and that the state has also seen Biden’s favorability rise post-South Carolina. Depending on how that support is distributed throughout Minnesota’s eight congressional districts, Sanders and Biden could take about an equal share of the state’s delegates, with both Warren and Bloomberg winning delegates too.
It’s Sanders who has the home-field advantage in Vermont, a state he is expected to win by a wide margin — Data for Progress’s Tuesday poll has his support in the state at 57 percent. The state only has 16 delegates, but victory seems pretty much assured for the senator.
Sanders also hopes to eke out wins in nearby states, including in Warren’s home state, Massachusetts — a state with 91 delegates. For much of the race, Warren was in the lead there, but that changed in February as Sanders’s campaign gained steam with wins in New Hampshire and Nevada.
A Suffolk University poll taken from February 26 to 29 has the two candidates neck-and-neck, with Sanders at 24 percent support and Warren at 22 percent. With a 4.4. percentage point margin of error, the poll suggests the race could still go either way, as does Data for Progress’s work, which found Warren to have a 2 percentage point lead on Sanders in a survey that has a 5.6 percentage point margin of error.
Should results reflect this polling, expect Sanders and Warren to walk away with a similar number of delegates, as they will have received about the same share of the vote.
Overall, Super Tuesday looks to be a promising day for both Sanders and Biden — perhaps especially for Biden, who just four days ago trailed one of his new surrogates, Buttigieg, in the delegate count. Warren is counting on winning broadly rather than big, and there are signs that strategy may bear fruit, although it does not appear it will be successful enough to turn her into a frontrunner overnight. Bloomberg similarly appears poised to reap some — if not a race-altering number — of delegates from his Super Tuesday strategy as well.
What the polls do not suggest, however, is a huge victory for any one candidate. Meaning the contests to come could be just as important as Tuesday’s.